Figure 3
From: Robots Learn to Recognize Individuals from Imitative Encounters with People and Avatars
Neural network (N.N.) activities during the learning phase in experiment 1.
The number of neurons needed to learn from the Visual Features N.N. (red); Prediction Error E(t) from the novelty detector N.N. (green); derivative of error prediction dE/dt (dark pink); average of dE/dt (black); novelty detection in the Person Recognition N.N. (dark blue); each hit corresponds to the detection of a novel interactive partner.
